Electricity Markets & Power System Operation in Australia Hugh Outhred, Presiding Director, CEEM Presented at the KPX International Workshop on Electricity Markets and Power System Operation, 5-7 July, 2005
Electricity Markets & Power System Operation in Australia
Hugh Outhred, Presiding Director, CEEMPresented at the KPX International Workshop on Electricity Markets and Power System Operation, 5-7 July, 2005
Outline
The process of electricity industry restructuringImplementation of restructuring in AustraliaLessons from the Australian experienceConclusions
Electricity industry restructuring:
Is like deciding to change from driving on the right hand side of the road to driving on the left hand side of the road. Without stopping the traffic.It is better not to do it at all unless:– You are sure you want to do it– You know what you want to do and how to get it done– You don’t stop halfway
Australia took its time to plan what to do & decide it wanted to do it. It then did the job sufficiently well.However, the outcome remains at risk
The electricity industry restructuring process: diversifying decisions, broadening options, spreading risks
Variable RE energy flowsEnd-user participation;Accountability
From direct costTo full costs
Regulatory Decision-making(environmental)
Multiple objectives;Measuring outcomes;Accountability
From rate of return ToIncentive Regulation
Regulatory Decision-making(economic)
Market power;Market design fidelity;Accountability
From cost recovery To market prices
CommercialDecision-making framework
Cultural change;Adequate competition;Accountability
From monopoly To competing firms Plus system operator(s)
Structural disaggregation
Key challengesTransitionIssue
Challenges in electricity industry restructuringUnderstanding & managing industry risks:– From short-term operation to long-term investment
Consistency in the decision-making framework:– From the short-term to the long-term future– From primary energy providers to end-users– Across the full scope of the electrical network
Decision-making compatibility:– Centralised: governance & regulation; system operation– Decentralised: participants as individuals and in groups
Decision-maker autonomy & accountability:– Participants, system operators, regulators, governments
Governance & institutions:- as seen by Allens Arthur Robinson, December 2003
Policy & regulatory context
Trading in electricity:- an abstraction from reality
Physical electricity industry(equipment; centralised; concrete)
Ancillaryservices;
Governance;Regulation(centralised)Engineering models
(equipment; centralised; abstract)
Commercial model: spot & derivative markets(humans; decentralised; abstract)
Economic models(humans; centralised; abstract)
Externalities
An electricity trading framework
transmissionnetwork
distributionnetwork
Interchange to other
wholesale market regions
WholesaleMarket region
distributionnetwork
distributionnetwork Retail
Market 3RetailMarket 1
Retail Market 2large consumer
• Wholesale & retail market designs should be compatible• Both should include network models
Primary energy markets
• Small consumers, embedded generators & storage should be supported by energy service advisers
risks to end-useenergy service
delivery
most consumers
large generators
embedded generators
Managing supply-demand balance in the Australian National Electricity Market
Spot market forecast &derivative markets
Frequency controlancillary service
markets for period tSupply/demand
projections & FCASderivative markets
FCAS marketsfor period t+1
Spot marketfor period t
Spot marketfor period t+1
time
spotperiod t
spotperiod t+1
increasing uncertainty
Physical issues
Commercial issues
Scope of the NEM
• Queensland• New South Wales & ACT• Victoria• South Australia• Tasmania (DC link to the
mainland in 2006)NEM regions are indicated, and their boundaries need not be on state borders (e.g. two regions in NSW)
Basslink DC link MNSP (2006?)600MW short term rating (north)
Murraylink DC link, now regulated, formerly MNSP
Directlink DC link, currently MNSP
180
National Electricity Market market regions(Securing Australia’s Energy Future, 2004)
Electricity industry structure in SE Australia
Gen 1
Gen 2
Gen X
GenerationSector:-
largegenerators
Gen 3
TransmissionSector
NSWVictoria
South Aust.Queensland& possiblyTasmania
TransmissionSector
NSWVictoria
South Aust.Queensland& possiblyTasmania
Electricity
Financial instrument& REC (emission) trading
Distributor 1
Distributor 2
Distributor Y
Distributionsector
Electricity
Multi-regionNational
ElectricityMarket(NEM)
Intentionsoffers &
paymentsRetailer Z
Retailer 2
Retailer 1
Retailsector
Intentionsbids &
payments
Tx networkpricing
Tx networkpricing
Networkaccess End-use
Equipment&
DistributedresourcesElectricity
End-usesector
Contestableend-users
FranchiseEnd-users
RetailMarkets
Embeddedgenerators
Average weekly price & demand, SA, Q1 2005(NECA, 05Q1 Stats, 2005)
Spot price duration curve, SA, Jan-Mar 05 (NECA, 04Q4 Stats, 2005; half-hour spot prices)
Weekly avg. NEM spot prices since market inception (NECA, 05Q1 Stats, 2005)
Running weekly accumulation of (336) RRPs & cumulative price threshold (CPT) (NECA, 05Q1Stats, 2005)
S10,000/MWH price cap reduced if CPT is reached
Distribution of NEM spot prices & revenues(Federal Government: Securing Australia’s Energy Future, 2004)
Flat contract prices, Q1 2005(NECA, 05Q1 Statistics, 2005)
D-cyhpaTrade exchange-traded call options for NSW peak period (www.d-cyphatrade.com.au)
0200400600800
100012001400160018002000
65 70 75 80 85 90Strike price $/MWH
05Q106Q107Q1
NEMMCO Statement of Opportunities, 2004
Changing generation offer to raise spot market price (2/8/03) graph courtesy of Stuart Thorncraft &Intelligent Energy Systems EMIS facility (www.iesys.com.au)(possible demand-side responses: derivative contract or reduce demand)
Region Supply Curve (NSW1)(02/08/2003 - 02/08/2003)
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
$/MWh
Region Supply Curve (NSW1)(02/08/2003 - 02/08/2003)
3500 4000 4500 5000 5500 6000 6500 7000 7500 8000 8500 9000 9500
MW
Delta Electricity Sithe Redbank Macquarie Generation
Eraring Energy
Price_30Min (NSW1)
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
10000
11000
MWPrice_30Min (NSW1)
Demand
Price
0
10
20
30
40
50
60
70
80
90
100
110
120
$/MWh
00 02 04 06 08 10 12 14 16 18 20 22Sat, 02 Aug 03
10:30
16:30
18:30
~95 $/MWH
~600MW
Robustness example:Transformer failure at 2142, 13 Aug 04:3,100MW gen lost;Frequency 48.9Hz; 2,100 MW load shed in NSW, Qld, Vic, SA(www.nemmco.com.au)
~900MW
~600MW
Robustness example: NEM energy revenue, 13/8/04
Robustness example: NEM FCAS revenue 13/8/04
Conclusions on the Australian experienceSuccessful electricity restructuring required:– Care in developing & maintaining consensus– A high level of professionalism in key roles– Extensive peer-review, auditing & testing of market design
Experimental economic techniques used to test design
Australian NEM uses a coherent design approach:– Decision making & risk management framework – Applied consistently to engineering & commercial issues,
recognising strengths & weaknesses of each– Design works successfully with a weak network, weather-
sensitive demand & a growing level of wind generation
1995 NEM rules trading experimentsImplemented key features of the NEM rules:– NEM bid and offer structure & network loss functions– Multi-region spot market with embedded network model– 1 & 2 day ahead STFMs with embedded network model– Short term PASA with broadcast warnings of constraints
Provided a sufficiently realistic trading environment:– Key participant characteristics– Uncertainty in availability, weather– Reporting on trading activity & simple decision support
Operated faster than real time:– 2 weeks of trading per day with 3 hour spot market period
NEM rules use a half-hour spot market interval
Used for education as well as formal experiments:– >1,000 participants prior to actual market start
Technology types in ‘95 NEM simulation modelSeven thermal generator portfolios (subject to outages):– Fast start generators (instantaneous start)– Slow start generators:- all other thermal plant (self commitment)
Max & min load; hot, warm & cold start-up timesStart-up costs; fixed & variable operating costsMean time to fail, mean time to repair, cost of repair
One hydro generator:– Headpond capacity & initial level, inflow rate, pumping
Seven demand side participants (uncertain temperature):– Retail tariff load (daily & weekly patterns, temp. coef’t)– Demand management (psuedo generator with op. cost)
One reserve participant (small thermal plant)
Simulation results: single region pricesPrices S2311
05
101520253035404550
Wed
03W
ed15
Thu0
3Th
u15
Fri0
3Fr
i15
Sat0
3Sa
t15
Sun0
3Su
n15
Mon
03M
on15
Tue0
3Tu
e15
Wed
03W
ed15
Thu0
3Th
u15
Fri0
3Fr
i15
Sat0
3Sa
t15
Sun0
3Su
n15
Mon
03M
on15
Tue0
3Tu
e15
$/M
Wh Spot Price
STFM1 PriceSTFM2 Price
Occasional zero spot prices. After the initialisation phase, theSTFM prices provide reasonable predictions of average spot prices
Simulation results: single region quantitiesSummary MW S2311
-10000
0
10000
20000
30000
40000
50000
Wed
03W
ed15
Thu0
3Th
u15
Fri0
3Fr
i15
Sat0
3Sa
t15
Sun0
3Su
n15
Mon
03M
on15
Tue0
3Tu
e15
Wed
03W
ed15
Thu0
3Th
u15
Fri0
3Fr
i15
Sat0
3Sa
t15
Sun0
3Su
n15
Mon
03M
on15
Tue0
3Tu
e15
MW
Contract CoverCommitted MWDemand
Contract cover is a compromise between generator interests and demand side interests. Some contract speculation evident on the final Monday. Generators de-committing at weekends but not overnight.
Outcomes from trading experimentsA sophisticated simulation package can adequately represent NEM market trading conditions:– Operational focus; two weeks of simulated trading per day– Suitable for exploring operation of market rules & for training
Experiments demonstrate that generators can exercise market power if insufficient competition:– Only under outage conditions for single region experiments:
Largest generation participant then had ‘must run’ plant– Most of the time in each region for five region experiments:
Single or larger generation participant in each region had ‘must run’ plant because of tie-line flow constraints
– Forward contract cover can protect a buyer from the effectsof spot market power
Key references (these & other publications at www.ceem.unsw.edu.au):1. H R Outhred & R J Kaye (1996), “Incorporating Network Effects in a Competitive Electricity Industry: An Australian Perspective”, Chapter 9 in M Einhorn & R Siddiqi (eds), Electricity Transmission Pricingand Technology, Kluwer Academic Publishers.2. H R Outhred (2003), “Some Strengths and Weaknesses of Electricity Industry Restructuring in Australia”, Paper 235, Proceedings of IEEE PowerTech ’03 Conference, Bologna, 23-26 June, ISBN 0-7803-7968-3.3. H R Outhred (2004), “The Evolving Australian National Electricity Market: An Assessment” in Power Progress: An Audit of Australia’s Electricity Reform Experiment edited by Graeme Hodge, Valarie Sands, David Hayward and David Scott, Australian Scholarly Publishing, Melbourne,ISBN 1 74097 034 9, 2004.